Project: Spectrum | Pilot Production Run Assembly, Testers, And Firmware


Beloved community,

What an exciting week! Right from the start, I captured the pilot production run assembly, and on the second day, post-assembly line testing after the newly assembled units completed their overnight burn-in. When I get time to sit down and light up the Spectrum prototype that’s with me day and night (even follows me to-and-from the factory as I love communicating Spectrum firmware on Spectrum), I catch up on the super energetic tester chat and guide this secret team on detailing all findings for our firmware team to improve Spectrum. A plentiful supply of real-world use testing from our ever-helpful testers not only helps me precisely track the current firmware behavior, but inspires me to delve deeper into existing functions and think out of the box to propose solutions that make them even better.

Production starts

Here’s a sneak peek fresh from the factory before we introduce this process in-depth with our upcoming topic:

Individual prototype impressions

If you haven’t done so, definitely check out our Project: Spectrum | Prototypes category, where many of our testers already released a handful of their unique hands-on experiences. This ranges from high-end PC gaming, HDR video playback, coding, KVM switch, overdrive, to backlight dimming. If you are interested in asking our lovely testers to try out your use cases, don’t hesitate to ask questions on their topics!

Our unusual ‘QRE’ team

Behind the scenes, the team is working closely with our testers on the continuous refinement of our firmware and OSD. Similar to our quality and reliability engineering (QRE) professional testing team, our testers co-work on a document to keep all findings and suggestions in regular check and update their status as new firmware gets released. Our firmware team reaches test data from an even broader range of user devices through maintaining this tracker and subsequently optimizing Spectrum with much-improved efficiency.

Challenge the status quo

Our tried and true user-upgradeable firmware enables us to look into the endless possibility of further refinement upon existing features with a simple firmware file delivery later down the line. We are redefining our 100 steps of brightness adjustment to bring the minimum brightness even lower for better viewing comfort in a dark environment and more evenly spaced brightness steps for a more granular control required by specific color-critical works. Moreover, the best overdrive settings are being explored based on four separate 64-step tables. Based on the fact that response time increases as the user defined overdrive level is moved up, we start by capturing and analyzing perceived overshoot levels to identify all meaningful user defined overdrive levels as potential candidates to become more refined Normal and High setting. After pinpointing all the settings we are interested in, customized response time measurement data are obtained with the help of our manufacturer to determine the best table to use for Spectrum’s user defined overdrive, and within the same table, the best values for Normal and High settings.

For comparing perceived overshoot and clarity levels, we currently use a fixed position camera. Although it may not be ideal for comparing adjacent levels of the same table, setting it to an appropriate shutter speed can distinctly indicate the increase of overshoot level and clarity as user defined overdrive level increases. It also helps observe similar overshoot levels across settings of different overdrive tables.

Test version firmware; overdrive table 4; shutter speed: 1/30s
Overdrive off
User defined level 18, Normal candidate
User defined level 41, High candidate

As Spectrum production is quickly moving forward, we are looking forward to bringing you more progress updates soon! Stay tuned!



It’s really exciting to see some Spectrums rolling off the production line, even if it is a pilot run.

Hopefully the worst of the firmware issues are fixed in time to avoid a day 1 update for customers. It sounds like HDR is causing a few headaches, but everything else is looking really good.


Looks so good, cant wait to get mine!
One thing caught my eye; at 0:37 we get to see the DeltaE, looks like an average of 0.5 and a max of 3.74
The reds seem to be the outliers by a good bunch?
So a couple questions:

  1. Does this mean the monitor will be slightly “warm” or just that the reds/skin tones will be slightly inaccurate? Red is a pretty important color, i am a bit concerned of the high DeltaE value there
  2. Is this pretty much the expected end calibration, or will this improve over time or another batch or something else?
  3. Is the DeltaE expected to be similar b/w the 3 models?

I think that is due to the red phosphor backlight that the LG panels use.

@Lore_Wonder , I’d love to have a real-time updated table with the current status of all the Todos (Features/Bugs/Issues). It gets harder to keep track of all the bugs and issues and their current status. Something like this (see screenshot) would really help a lot of us.

Is it something doable for you guys?


Keep these updates coming @Lore_Wonder
We’re entering the most exciting stage of this project and I am sure everyone is on the edge of their seat to see Eve coming trough the final meters towards shipping out the product.


Thank you, guys! I’m delighted to hear it! :grin: :tada:

Good color out of the box is our focus at the moment, and improving the HDR experience before shipping is a crucial step to achieve this goal. Our firmware team is hard at work ironing out each and every misbehavior our testers experienced.

We are currently investigating the root cause of this reported red Delta E value, including using independent color measurement tools to cross-check the behavior of color 255, 0, 0 (RGB) on factory-calibrated prototypes.

Colorimeter 255, 0, 0

Later we’ll be back with an update on our progress!

It’s not the expected calibration result with the presence of the two outliers. This should get improved with our progress on the previous question.

Most likely, the Delta E will be similar among the three models. We will test it out when we have the appropriate prototypes.

Firmware issues discovered are appropriately debugged by our firmware team and get retested to make sure all of them are eliminated eventually. On prototypes, they always appear much worse than they are in reality, and can be misinterpreted. Therefore, this table is not something we are willing to provide in the community. Regardless, if you are curious about how Spectrum prototypes handle your use cases, don’t hesitate to leave a comment to the team or our ever-helpful testers and let us experience it for you!